Processing drivable surfaces for simulated environment systems and applications

Abstract

In various examples, one or more of the embodiments apply a path elevation smoothing process to navigable path surface data to derive road surface rendering data that may be used to define a drivable surface within a simulated driving environment. In some embodiments, a surface smoothing process computes smoothing correction coefficients that may be applied to a baseline road surface map. Using the smoothing correction coefficients, a plurality of points from the baseline road surface map may be projected to render a smoothed surface that may be used to generate the drivable surface. In some embodiments, smoothing correction coefficients may each comprise a directional correction component that smooths series-connected lanelets in the direction of travel of the road surface to produce a set of lane ribbons. The smoothing correction coefficients may each further comprise a cross-directional correction component that smooths across the set of lane ribbons.

What is claimed is: 
     
         1 . A system comprising one or more processors to:
 generate a road surface map based on connecting two or more individual representations of adjacent sections of a roadway surface represented by road data, wherein the road surface map comprises a topological mesh of interconnected vertices;   generate a plurality of lane ribbons based on smoothing, with respect to elevation, the road data along a first axis;   generate a plurality of curves based on smoothing, with respect to elevation, the plurality of lane ribbons along a second axis;   compute a set of smoothing values based at least on a displacement between at least one curve of the plurality of curves and individual vertices of the topological mesh of interconnected vertices; and   generate a simulated road surface within a simulated environment based on applying the set of smoothing values to the road surface map.   
     
     
         2 . The system of  claim 1 , wherein for at least a first segment of the roadway surface, the first axis is aligned with a direction of vehicle travel associated with the roadway surface, and the second axis is aligned perpendicularly to the first axis. 
     
     
         3 . The system of  claim 1 , wherein the road data is derived at least based on LIDAR data representing the roadway surface. 
     
     
         4 . The system of  claim 1 , wherein the one or more processors are further to compute the set of smoothing values based at least on:
 a first set of correction components computed based at least on one or more displacements between the individual vertices and one or more lane ribbons of the plurality of lane ribbons; and   a second set of correction components computed based at least on one or more displacements between one or more lane ribbons of the plurality of lane ribbons and one or more curves of the plurality of curves.   
     
     
         5 . The system of  claim 1 , wherein the one or more processors are further to generate the simulated road surface based on projecting the individual vertices from the road surface map using the set of smoothing values to define one or more elevations for the simulated road surface. 
     
     
         6 . The system of  claim 1 , wherein the one or more processors are further to store individual smoothing values as correction data correlated to at least one of the individual vertices of the topological mesh. 
     
     
         7 . The system of  claim 1 , wherein the one or more processors are further to:
 generate the plurality of lane ribbons further based at least on alighting a lane boundary between neighboring lane ribbons.   
     
     
         8 . The system of  claim 1 , wherein the one or more processors are further to:
 generate one or more supplemental lane ribbons based on overlapping representations of the roadway surface from the road data; and   adjust the set of smoothing values based on the one or more supplemental lane ribbons.   
     
     
         9 . The system of  claim 1 , wherein the one or more processors are further to generate at least one of the plurality of lane ribbons or the plurality of curves by executing a smoothing algorithm based on at least one of: bin smoothing, kernel smoothing, simple moving average, local weighted regression, or parabola fitting. 
     
     
         10 . The system of  claim 1 , wherein the one or more processors are further to:
 connect at least one roadway intersection within the simulated environment with the simulated road surface based at least on aligning an elevation of the road surface map with a baseline intersection surface generated from connecting individual representations of adjacent sections of the at least one roadway intersection.   
     
     
         11 . The system of  claim 1 , wherein the one or more processors are further to generate the plurality of lane ribbons based at least on connecting one or more roadway intersections. 
     
     
         12 . The system of  claim 11 , wherein the one or more processors are further to:
 adjust at least one value of the one or more smoothing values for the simulated road surface based on blending one or more surface elevations of the simulated road surface with one or more surface elevations of the one or more roadway intersections determined by an intersection smoothing process.   
     
     
         13 . The system of  claim 11 , wherein the one or more processors are further to:
 compute one or more surface elevations of the one or more roadway intersections using an intersection smoothing process based at least on one or more surface elevations of the simulated road surface.   
     
     
         14 . The system of  claim 1 , wherein the one or more processors are comprised in at least one of:
 a control system for an autonomous or semi-autonomous machine;   a perception system for an autonomous or semi-autonomous machine;   a system for performing simulation operations;   a system for performing digital twin operations;   a system for performing light transport simulation;   a system for performing collaborative content creation for three-dimensional assets;   a system for generating or presenting at least one of virtual reality content, augmented reality content, or mixed reality content;   a system for performing deep learning operations;   a system for performing real-time streaming;   a system implemented using an edge device;   a system implemented using a robot;   a system for performing conversational AI operations;   a system for generating synthetic data;   a system incorporating one or more virtual machines (VMs);   a system implemented at least partially in a data center;   a system for performing generative AI operations;   a system implemented at least partially using a language model; or   a system implemented at least partially using cloud computing resources.   
     
     
         15 . One or more processors comprising processing circuitry to:
 generate a road surface map based on connecting individual representations of one or more adjacent sections of a roadway surface;   compute a set of smoothing values based at least on a plurality of lane ribbons and a plurality of curves, the plurality of lane ribbons being generated by smoothing the road surface map with respect to elevation along a first axis of the roadway surface and the plurality of curves being generated by smoothing the plurality of lane ribbons with respect to elevation along a second axis; and   generate a simulated road surface within a simulated environment based on applying the set of smoothing values to the road surface map.   
     
     
         16 . The one or more processors of  claim 15 , wherein the road surface map comprises a topological mesh of interconnected vertices, wherein the processing circuitry is further to compute the set of smoothing values based at least on a displacement between the plurality of curves and individual vertices of the topological mesh of interconnected vertices. 
     
     
         17 . The one or more processors of  claim 16 , wherein the one or more processors are further to compute the set of smoothing values based at least on:
 a first set of correction components computed based at least on one or more displacements between the individual vertices and the plurality of lane ribbons; and   a second set of correction components computed based at least on one or more displacements between the plurality of lane ribbons and the plurality of curves.   
     
     
         18 . The one or more processors of  claim 15 , wherein the one or more processors are further to generate the plurality of lane ribbons based at least on connecting one or more roadway intersections within the simulated environment. 
     
     
         19 . The one or more processors of  claim 15 , wherein the one or more processors are comprised in at least one of:
 a control system for an autonomous or semi-autonomous machine;   a perception system for an autonomous or semi-autonomous machine;   a system for performing simulation operations;   a system for performing digital twin operations;   a system for performing light transport simulation;   a system for performing collaborative content creation for three-dimensional assets;   a system for generating or presenting at least one of virtual reality content, augmented reality content, or mixed reality content;   a system for performing deep learning operations;   a system for performing real-time streaming;   a system implemented using an edge device;   a system implemented using a robot;   a system for performing conversational AI operations;   a system for generating synthetic data;   a system incorporating one or more virtual machines (VMs);   a system implemented at least partially in a data center;   a system for performing generative AI operations;   a system implemented at least partially using a language model; or   a system implemented at least partially using cloud computing resources.   
     
     
         20 . A method comprising:
 generating a surface for rendering a drivable area within a simulated driving environment based on computing a set of smoothing values describing a displacement between a baseline road surface and a plurality of curves generated by smoothing a plurality of lane ribbons with respect to elevation, the plurality of lane ribbons being generated by smoothing the baseline road surface with respect to elevation, wherein the baseline road surface is generated by connecting individual representations of adjacent sections of a roadway surface.